What a Corporate Livestream Costs: $4,000 to $20,000

The question arrives in almost every first call. There is no rate card to answer it with. There is a shape, though, and it is worth understanding before you brief anybody, because the cost of a livestream does not rise smoothly: it steps up as new gear and labor are required.

Most productions we run land between $8,000 and $20,000 per day. Smaller single-camera shows start around $4,000, and below that we are not the right call and will tell you so. That is the range. What follows is why a given show sits where it sits inside it.

What does a single-camera livestream get you?

One camera, one operator, an encoder. The camera sees the room, the encoder sends it out, and that is the entire signal path. That is the $4,000 end of the range, and it is the one rung with no graphics on it at all.

What it buys is a faithful record. What it cannot do is put anything on the screen. No name key under the speaker, no opening card, no cut to a slide, no holding graphic when something goes wrong. Nothing in that chain is capable of composing an image, so the picture is whatever the camera is pointed at, for the length of the show.

That is a real product and it is the right one more often than people expect. A single speaker, a ceremony, a service, a hearing, a class: anywhere the value is the record rather than the presentation of it.

Why a lower third and a second camera both need a switcher

Add a switcher, and the operator to run it, and the screen becomes something you can compose. A lower third with somebody’s name. An opening card and a closing card. A slide feed. Something to cut to when a problem needs thirty seconds of cover.

The part worth understanding is that two completely different asks push a show across the same step:

• You want graphics, even one name key.
• You want a second camera.

Neither is possible without a switcher, so both requests run into the same wall. Can we just add a lower third and can we just add a second angle turn out to be the same question.

The consolation is that the wall gets paid for once. What sits past it is not the same for both, though. A holding card or a name key needs the switcher and nothing else, so once the step is paid, simple graphics are close to free. A second camera needs the switcher and then a camera, and a mount or somebody to point it.

From there the two asks diverge on their own terms. Cameras stay comparatively cheap, because the expensive part was the switcher and the person running it, and each camera after the first adds roughly its own cost. Graphics climb in their own direction: one card is nothing once the step is paid, and a full animated package is a build with somebody to run it.

What separates an $8,000 show from a $20,000 one

This is the real answer to why one livestream costs twice what another does. Neither half of a quote is a mystery: gear is specified line by line, crew is specified the same way, and both are predictable before anybody signs anything. Labor usually runs a bit ahead of gear on the total, and it is the side that opens the distance between the bottom and the top of our range, because what changes as a show grows is how many of the jobs below are separate human beings rather than one person covering several.

Concretely, the two ends of that range look like this:

• Around $8,000: three PTZ cameras run by one person from the control position, a simple graphics package, a simple sound package of two or three lavaliers, and a single RTMP feed going out.
• Around $20,000: three or more operated cameras with an operator each, a second program build (vertical as well as horizontal, or an LED wall in the room to feed), and sound that has outgrown a set-and-leave plan.

Almost everything in that gap is a person or a second build rather than a more expensive box.

The camera decision shows it most clearly. Four operated cameras means four operators. Four PTZ robotic cameras can be run by one person from the control position, which is exactly why robos turn up in tight rooms and tighter budgets. The saving is real, and so is the trade: one person riding four cameras is not composing four shots the way four operators would.

Who is on a livestream production crew?

On a small show one or two people cover all of this. As the show grows, each of these stops being something anybody can do on the side:

• Production assistants. The lowest cost line on the list and the one the rest of it stands on. They run cable, wrangle talent, keep the room moving, and go and get whatever the show turns out to need. A bigger show does not need a different kind of PA, it needs more of them.
• Camera operators. One per operated camera, or one across several robotic ones.
• A sound engineer. Once the audio is more than a single feed you can set and leave, somebody has to be mixing it while the show happens.
• A graphics operator. An opening and closing card can be loaded and cut to by the technical director. Sixty-five lower thirds cannot, and an animated package is a separate build with a timeline of its own.
• A remote guest tech. Somebody whose entire job is the people who are not in the room: bringing them in, keeping their return feed and mix-minus correct, and handling it when their connection degrades mid-answer.
• Lighting. If the room’s own light is the light, that is free, and it is usually the reason a stream reads as amateur even when everything else was done well.
• An engineer in charge. Owns the technical flow end to end, signal and network and encode and delivery, and owns the decision when a piece of it stops.
• A technical director. Cuts the cameras and the graphics into the program feed.
• A show caller. Calls what happens next and when: shots, cues, transitions, playback.

At the small end the last three on that list are one person. The same operator owns the technical flow, cuts the program and calls the show, and that holds because none of the three is asking for full attention yet.

What separates them is not really the size of the show. It is how much is happening on cue, and how many systems are running. Once more than three people are executing cues (camera cuts, graphics, lighting, sound, etc.), a show caller is probably warranted, because somebody has to be deciding what comes next while everybody else is executing.

The engineer in charge separates on a different axis, and it is not cue volume. A hundred graphics cues is a busy graphics operator, not a reason to add an engineer. What calls for one is the number of independent systems that have to work together. A graphics system, an LED wall, a playback system and a downstage monitor are four separate things that all have to agree with each other, and somebody has to own the space between them. Each separation is another person on the quote, which is most of what moves a show up the range.

Which role do clients want to cut first?

It is the engineer in charge, near enough every time, and for an understandable reason: when the show goes well that person appears to have done nothing at all. That is the job working. The difference between a problem solved in ninety seconds and a problem that takes the broadcast down is almost always whether somebody who understood the whole system was already watching when it started. On a complex show we will not take it out of the plan. A show with enough moving parts to need an engineer in charge is exactly the show that cannot absorb not having one.

Why a one-hour webcast can be a two-day job

The other reliable underestimate. A one-hour webcast is not a one-hour job.

Building a live television studio inside your office takes between two hours and a full day, depending on the show. A technical rehearsal happens the day before, not the hour before. So an event that is on air for sixty minutes routinely occupies two days of crew and equipment, and the quote reflects that rather than the runtime.

It is also why late additions cost more than they look like they should. The gear was specified weeks ago and the crew was booked weeks ago. Adding a camera at T minus two is not the price of a camera. It is the price of a camera at short notice, plus whoever is now needed to point it.

Here is what that looks like when it happens. A client told us they would have four remote guests, so we quoted a remote guest system that handles four at a time. On the build and rehearsal day, which is the day before going live, they asked for a fifth. Our eight-guest hardware was already booked on another show.

We sent a PA to B&H, bought new components, took the four-guest system apart and rebuilt it around them. They got their fifth guest. What they did not get was certainty, because until late in that day nobody could promise it was going to work. One more guest sounds like one more window on a screen. It was a hardware capacity limit, set weeks earlier when the system was specified, and the only reason it was solvable at all was that the ask landed on a build day, with a shop still open and somebody free to go to it.

Which livestream upgrades are not worth the money?

Some of what gets asked for, we will talk you out of:

• More cameras than the stage warrants. Three to six earns its keep on a four-person panel where the conversation moves between people. On one speaker at a lectern it is money spent on angles nobody will ever cut to.
• Battery backup beyond a graceful shutdown. If the building loses power, the stage lighting and the LED wall go with it and the show is over regardless of how much runtime our rack has left. Unless there is a generator capable of keeping the HVAC running, buy enough to close the recordings cleanly and stop there.
• Movement for its own sake. A jib, a slider or a Steadicam is worth it when the shot it makes is worth it. Otherwise it is a rental line and an operator to go with it.
• A sound stage when the room is the point. A stage removes problems, no HVAC noise, no windows, no room tone, and it also removes the event from the place where it means something. Decide that deliberately rather than by default.

And the honest one. Below about $4,000 we are not the right vendor. That is not a negotiating position, it is the point beneath which we cannot put a crew in a room and do work we would want our name on, and we would rather point you to someone who fits.

What we need to know before quoting a show

Every variable above is set by what is happening on stage and who needs to see it, which is why the useful conversation does not start with the budget. Tell us the room, the largest number of people who will be on stage at once, whether the audience asks questions, where the stream needs to go, and roughly how much has to appear on screen. That is enough to size the job, and the number falls out of it.

One last thing, because it is the most common way to waste money on this curve: do not cross a step and then decline to use it. If a show is already paying for a switcher and somebody to run it, the holding card, the slide feed and the opening and closing bookends need nothing beyond it and are close to free. A second camera is not free, but it is cheap set against what the step already cost. Cross the step deliberately, or stay well below it. Buying half of it is the worst value on offer.